A 33 location, 50 hybrid test was evaluated for the planting year 2004 to determine the optimal number of one replication locations to accurately identify the best hybrids to move forward in a corn breeding program. The 33 locations were located in maturity zone seven and balanced to have equal representation across the zone from east to west. Hybrids were created from fixed inbred lines. For the purposes of this experiment best hybrids were defined as those having the highest mean yield across all locations. One-thousand iteration bootstrap analyses were performed, randomly selecting from four to 15 locations (without replacement). Percent hybrids retained for the analysis were 5, 10 and 15 percent of the hybrids tested and the frequency of occurrence of the top three hybrids as defined above were evaluated. Average yield by site was 229 bushels and ranged from 179 to 263 bushels. When evaluating four locations, the best and third best hybrid were selected 50% and 28% of the time, respectively, when the top 5% percent of hybrids were retained. The best and third best hybrid were selected 74% and 62% of the time, respectively, when the top 15% of hybrids were retained. When evaluating eight locations, the best and third best hybrid were selected 71% and 37% of the time, respectively, when the top 5% percent of hybrids were retained. The best and third best hybrid were selected 93% and 79% of the time, respectively, when the top 15% of hybrids were retained. Based on the results of this study and supporting literature, a one replication, four location yield trial, retaining the top five percent of selections was too great a risk; however, relaxing the retention rate to 15 percent was adequate to achieve the goals.